DRONE-BASED IMAGE ANALYSIS FOR EARLY DETECTION OF STRUCTURAL CRACKS IN AGING BRIDGES

Abstract

Aging bridges are increasingly vulnerable to structural cracking due to material deterioration, repeated traffic loading, environmental exposure, and delayed maintenance. Traditional bridge inspection methods are often time-consuming, labor-intensive, costly, and limited by accessibility challenges, especially for high-span, remote, or heavily trafficked bridges. This paper presents a drone-based image analysis approach for improving the early detection of structural cracks in aging bridges. High-resolution images captured through unmanned aerial vehicles were analyzed using computer vision and image-processing techniques to identify visible crack patterns, classify crack severity, and support timely maintenance decisions. The results indicate that drone-assisted inspection improves detection coverage, reduces inspection time, and enhances the identification of fine surface cracks compared with manual visual assessment. As Fig. 1 shows, the proposed model achieved strong performance across accuracy, precision, recall, and F1-score, while Table 1 confirms consistent detection across multiple bridge components. The findings further demonstrate that image quality, lighting conditions, flight altitude, and crack width significantly affect detection reliability. Overall, the study highlights the potential of drone-based structural monitoring as a practical, scalable, and cost-effective solution for early crack detection in bridge maintenance systems. The proposed approach can assist engineers and infrastructure authorities in prioritizing repairs, reducing inspection risk, and extending the service life of aging bridge assets.